正态分布下的最大似然
import numpy as np
from scipy.stats import norm
from scipy.optimize import minimize
import matplotlib.pyplot as plt
from matplotlib.font_manager import FontProperties
import matplotlib as mpl
mpl.rcParams['font.family'] = 'SimHei'
font = FontProperties(fname=r"/usr/share/fonts/truetype/arphic/uming.ttc", size=14)
mean = 0.9
std_dev = 1
sample_size = 100
random_data = np.random.normal(mean, std_dev, sample_size)
def neg_log_likelihood(params, data):
mean, std_dev = params
pdf_values = norm.pdf(data, loc=mean, scale=std_dev)
neg_log_likelihood_value = -np.sum(np.log(pdf_values))
return neg_log_likelihood_value
initial_guess = [0, 1]
result = minimize(neg_log_likelihood, initial_guess, args=(random_data,))
mean_mle, std_dev_mle = result.x
print(f"最大似然估计的均值为: {mean_mle}")
print(f"最大似然估计的标准差为: {std_dev_mle}")
plt.hist(random_data, bins=20, density=True, alpha=0.6, color='b', label='随机数据')
plt.hist(random_data, bins=20, density=True, alpha=0.6, color='b', label='随机数据')
x = np.linspace(min(random_data), max(random_data), 100)
y = norm.pdf(x, loc=mean_mle, scale=std_dev_mle)
plt.plot(x, y, 'r-', lw=2, label='拟合的正态分布')
plt.legend()
plt.xlabel('值')
plt.ylabel('密度')
plt.title('拟合的正态分布')
plt.show()